An Analysis of Questioning and Feedback Strategies Using the IRF Framework Classroom Research and Research Methods Paper submitted July 2010 to the School of Humanities of the University of Birmingham, UK in part fulfillment of the requirements for the degree of Master of Arts in Teaching English as a Foreign or Second Language (TEFL/TESL) LT/10/08 For an EL class to which you have access, investigate both of the following questions. (a) What are the teacher’s questioning strategies, and how effective do these appear to be? (b) What are the teacher’s feedback strategies, and how effective do these appear to be? Write a report of your findings.
28
Embed
An Analysis of Questioning and Feedback Strategies Using ... · PDF fileAn Analysis of Questioning and Feedback Strategies Using the ... and tally sheets were ... An Analysis of Questioning
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
An Analysis of Questioning and Feedback Strategies Using the IRF
Framework
Classroom Research and Research Methods
Paper submitted July 2010
to the School of Humanities of the University of Birmingham, UK
in part fulfillment of the requirements for the degree of
Master of Arts
in
Teaching English as a Foreign or Second Language (TEFL/TESL)
LT/10/08
For an EL class to which you have access, investigate both of the following
questions.
(a) What are the teacher’s questioning strategies, and how effective do these
appear to be?
(b) What are the teacher’s feedback strategies, and how effective do these
appear to be?
Write a report of your findings.
1
Centre for English Language Studies
Postgraduate programmes, Open Distance Learning
ESSAY COVER SHEET AND DECLARATION
Student ID number 1119155
Module Number (1-6) 1
Title of Degree Programme: MA TEFL / TESL
Title of Module: Language Teaching and Classroom Research Methods
Assessment Task No. LT/10/08
Submission First Submission
Date Submitted 10th July 2010
Name of tutor Gregory Hadley
Word Count 4,245
2
1. Introduction
Asking questions and giving feedback are activities which are central to the role of teaching.
There is an abundance of research in both areas, grouping questions into distinct categories;
as well as analyzing feedback in order to ascertain both its purpose and its effectiveness
(Chauldron, 1977; Ellis, et al., 2001; Guan Eng Ho, 2005; Havranek, 2002; Lyster and Ranta,
1997; Seedhouse, 1996). It is commonly accepted that teachers should become aware of the
types of questions and feedback they use if they are to excel as teachers.
In this paper I will explore the types of questions I ask and analyze feedback types in my
classroom. I will also examine my strategies and determine how effective they are according
to our current understanding of questions and feedback.
3
2. Literature Review
In this section, I will outline the IRF cycle, and explain how questions and feedback are
intertwined in this cycle. I will also discuss where key players in ELT stand on the issues of
questions and feedback.
2.1 The IRF Cycle
In language teaching, the IRF exchange (sometimes referred to as IRE) is the familiar
sequence of teacher-student-teacher turn-taking in the classroom. In the “initiation” (I) phase
the teacher usually asks a question, to which the student responds (R). This is then followed
by some sort of feedback or evaluation (F/E) by the teacher.
According to Van Lier, depending which questions are asked, the initiation stage may
„…require students merely to recite previously learned items.‟ (2001: 94) On the other hand,
he goes on to say:
‘At the most demanding end of IRF, students must be articulate and precise; they are pushed by
successive probing questions, to clarify, substantiate, or illustrate a point that they made previously.’
(Van Lier, 2001: 94)
While the IRF format may not be inherently ineffective, it could be considered restrictive, in
that students aren‟t able to initiate themselves. (Van Lier, 95) It might be possible to conclude,
however, that this form of interaction could be viewed as more pedagogically sound if the
teacher were to ask more referential questions vs. display questions, which would give IRF
the purpose of scaffolding, as Van Lier (2001: 96) suggests:
‘The initiation-response-feedback exchange, at least when it moves beyond mere recitation and
display, can be regarded as a way of scaffolding instruction, a way of developing cognitive structures
in the zone of proximal development, or a way of assisting learners to express themselves with
maximum clarity.’
Perhaps he is advocating that we design our questions with consideration to Vygotsky‟s
(1978: cited in Van Lier, 2001: 96) “zone of proximal development.”
2.2 Questions
We can now focus on the first phase of this interaction pattern. Questions, broadly speaking,
can be classified into two categories: display (or closed) questions, in which the teacher
already knows the answer, and referential (or open) questions, or those to which the teacher
does not know the answer (Barnes, 1975; Long and Sato, 1983; Guan Eng Ho, 2005). It is
generally recognized that teachers ask predominantly display questions in the classroom
(Nunan, 1987).
4
Nunan (1987) said that referential questions should be used more often than display
questions if we are to have more genuine communication in the language classroom. His
stance is not in line with the traditional IRF sequence: „…in genuine communication,
decisions about who says what to whom and when are up for grabs.‟(Nunan, 137) This would
imply that display questions are not compatible with the aim of communicative competence.
Seedhouse (1996: 21), in contrast, states that „In the classroom…the core goal is learning or
education, and both the IRF cycle and display questions are interactions well suited to this
course goal.‟ He also says (19) that the IRF cycle „…is very noticeably present in a particular
discourse setting outside the classroom, namely, in the home in parent-child-interaction.‟ It
would appear Nunan and Seedhouse do not see eye-to-eye on the definition of “genuine
communication.” Seedhouse also cites Van Lier‟s take on Nunan‟s idea that more referential
questions in the classroom equals more genuine communication: „Van Lier (1988, cited in
Seedhouse, 1996: 20) …argues that there is little difference, in interactional terms, between a
display question and a referential question.‟ One may conclude that Seedhouse and Nunan
also disagree on whether display questions are pedagogically sound.
Guan Eng Ho has suggested that questions cannot so easily be dichotomized into categories
of “display / closed” or “referential / open:” „There are frequent instances…where questions
asked and initiated by teachers are neither closed nor open.‟ (2005: 301) She has proposed a
third category of questions, which falls between the traditional two categories previously
mentioned. Included in this new grouping are „general knowledge, vocabulary, and language
proficiency questions.‟ (2005: 303) Presumably, she is trying to tell us that it is not so easy to
classify a question based simply on whether or not the teacher already knows the answer. She
goes on to say (305):
‘…if we are to consider the intentions behind the question, we would think twice about labeling it as
a closed type and therefore of inferior quality. And yet, it is the observer’s perspective that is made
known rather than those of the participants in the interaction.’
It follows, therefore, that while Nunan tells us that display / closed questions are without
purpose in the CLT classroom, and Seedhouse says that any classroom interaction with the
goal of learning is purposeful, Guan Eng Ho seems to be suggesting that effective questions
are those which accomplish what the teacher hopes to achieve in the first place, regardless of
any taxonomy a particular observer has assigned them.
2.3 Feedback
Feedback is the final phase of the IRF interaction. The Longman Dictionary of Teaching and
Applied Linguistics defines feedback in the context of language teaching as „… comments or
other information that learners receive concerning their success on learning tasks or tests,
either from the teacher or other persons.‟ (Richards and Schmidt, 2002: 199)
5
One common practice is to vary one‟s stress and intonation when giving corrective feedback,
and Zamel (1981: 145) takes issue with this:
‘…repeating part of the student’s answer with exaggerated stress of rising intonation (a practice I
have observed in many ESL classes) does not provide explicit information.’
Feedback has further been subdivided into several “feedback types,” such as those listed by
Panova and Lyster (2002), including back channeling, repetition, and recasts.
Back channeling, according to the Longman Dictionary of Language Teaching and Applied
Linguistics, is „feedback given while someone is speaking‟ and includes „comments such as
uh, yeah¸ really,…and grunts that indicate success or failure in communication.‟ (Richards
and Schmidt, 2002: 199)
Repetition, according to Zamel (1981: 144), would be more effective if the teacher were to
incorporate new information into feedback:
‘The use of analogy and contrast is more effective than frequent repetition:
Teacher: What are you holding?
Student: I am holding some paper.
Teacher: Good. Some paper. Just like some sugar, some coffee. (Emphasis on some)
Among these three types of feedback, recasts are the most common (Donato, 1994; Doughty,
1994; Ellis et al., 2001; Loewen, 2002, 2004; Lyster and Ranta, 1997; Oliver, 2000; Panova
and Lyster, 2002; Seedhouse, 1997; cited in Loewen and Philp, 2006: 537) Recasts are,
according to the Longman Dictionary of Teaching and Applied Linguistics (446):
‘a type of negative feedback in which a more competent interlocutor rephrases an incorrect or
incomplete learner utterance by changing one or more sentence components while still referring to
its central meaning.’
According to Mackey (2006: 405), recasts are one type of „helpful interactional
(process)…which can supply corrective feedback letting learners know their utterances were
problematic.‟ Havranek (2002: 268) seems to agree with Mackey, and adds: „Corrective
feedback is most likely to be successful if the learner is able to provide the correct form when
he is alerted to the error.‟
In order to measure the effectiveness of feedback, researchers generally use the learner‟s
response to the feedback (“uptake”) as a gauge (Chauldron, 1977; Loewen, 2004; Lyster,
1998; Lyster and Ranta, 1997; cited in Loewen and Philp, 2006: 539).
Mackey and Gass (2005: 199) have devised a framework for determining both frequency and
type of feedback uptake. Their categories indicate how soon a correction was incorporated
into the learner‟s next utterance, as well as whether or not an opportunity was given for
uptake or if the learner completely ignored the instructor‟s feedback.
6
While Mackey and Gass gain insight by examining the student‟s uptake and are more explicit
in what makes feedback effective or ineffective, Zamel (1981) puts the focus on the teacher:
‘There are many…instances in which the feedback is inexplicit, unrelated to the ambiguity the student
may be experiencing, and therefore ineffective.’ (145)
On a final note, it is prudent to acknowledge Havranek‟s warning (2002: 256) that:
‘Establishing that a correction succeeded…would require long-term observation of the learner’s
production of the corrected structure while at the same time making sure that there is no further
input of the same structure, ruling out any other source of learning.’
In other words, while using uptake is useful in analyzing one‟s own feedback strategies at a
particular place and time, observation over a longer period would be more telling.
7
3. Method
This section gives a background of the participants and the institution. It also describes the
methods of data collection used in this research. Finally, the procedure for analyzing data is
presented.
3.1 Participants
This study consisted of six adult Korean males, ages 42 to 50, studying in a ten-week
intensive English program in South Korea. The recording was made of a “Communication”
class, using Interchange 3rd
Edition, Book 2. The students were all at the same level,
“Intermediate Low,” according to their incoming Oral Proficiency Interview (OPI). All
students had previously studied English grammar in high school as well as university.
Therefore, they already had a rudimentary understanding of the grammar being practiced. At
the time of the recording, they had been involved in the program for nine weeks, so a rapport
between the instructor and students had been established.
The instructor (myself) has more than seven years‟ teaching experience, primarily with adults.
I had been at the institution for one year at the time this recording was made.
3.2 Procedures
The research presented here is based on the first half of a two-hour class, which was digitally
recorded and later transcribed. The data was then coded, and tally sheets were used in order
to both investigate the amount of questions and feedback (quantitative), and to classify them
with intent to reveal their efficacy (qualitative).
I strive to employ the communicative approach, in which „students are expected to interact
primarily with each other rather than with the teacher, and correction of errors may
be…infrequent.‟ (Richards and Rodgers, 2001: 66) This approach necessitated several
omissions from the transcript due to a large percentage of class time being allotted to pair or
group work, in which I did not interfere unless deemed appropriate.
The data analyzed consisted of any teacher-student interaction related to questioning or
correcting. The IRF framework was particularly helpful in determining which data would be
useful in my investigation.
It should be noted that this research is restricted to my own classroom, a prime example of
what is known as “convenience,” or “opportunity” sampling, a sampling strategy which is
non-probabilistic (Dornyei, 2007: 98). Therefore, the implications discussed here are
extremely limited in scope and should not be taken to be true for all teachers and students in a
similar situation. Or, as Dornyei points out, „No matter how principled a non-probability
sample strives to be, the extent of generalizability in this type of sample is often negligible.‟
(2007: 99)
8
3.3 Analysis
The categories used to code the first set of data were taken from the question types discussed
by Nunan (1987), Seedhouse (1996), and Guan Eng Ho (2005). I first categorized all
pedagogical questions as “display,” or “referential.” I then ascertained the purpose of the
other questions I asked, and assigned them the following categories:
Other
Seeking clarification
Checking comprehension
Procedural
Using Guan Eng Ho‟s criterion for effectiveness, I classified my questions as “effective” if I
was able to get the answer I wanted during the “response” stage of the IRF cycle. If I had to
probe further with another question, I classified the first question as “ineffective.”
My corrective feedback classifications were adapted from the observation scheme designed
by Mackey and Gass (2005: 199). I assigned each third move in the IRF cycle one of the
following categories:
Recast
Metalinguistic explanation
Clarification request
Repetition
Nonverbal cue
Praise
Criticism
Back Channeling
Again, using the Mackey and Gass framework, and the notion of learner uptake presented by
Loewen and Philp, I concluded my feedback was effective if the student incorporated my
correction either into their next utterance or at some point later in the class. I decided my
feedback was ineffective if the student had no opportunity for feedback incorporation or if he
ignored my correction.
9
4. Discussion
In this section I will discuss my questioning strategies and their degree of effectiveness.
Subsequently, I will examine the feedback strategies used in my class and assess their
efficacy.
4.1 Types of questions in my classroom
Table One shows the types and frequency of questions asked in my class:
Display //////// 8
Referential ////// 6
Other // 2
Seeking clarification /// 3
Checking comprehension /// 3
Procedural / 1
Total 23 Table 1
As is typical in second language classrooms, display questions were prevalent. Surprisingly,
the quantity of referential questions asked was only slightly less than that of display questions.
If we take “display questions” to be strictly those to which the teacher already knows the
answer, then the following two questions would certainly fall into that category:
1. Can you give me a synonym or another way of saying this word?
2. What kind of situation?
However, if we use the third category proposed by Guan Eng Ho, we can label these
questions as “Other.”
The remaining questions were largely procedural, asking students to repeat themselves if I
didn‟t understand; checking their understanding of my instructions; and giving one
instruction in the form of a question.
4.2 Effectiveness of questions
Of sixteen pedagogical questions asked, nine were deemed “effective” and seven were
“ineffective.” I will now address several examples of these. I will put my expected response
in parentheses.
The first question was asked at the beginning of the class in order to introduce the lesson
topic:
10
T: Predicament. Can you give me a synonym or another way of saying this word?
(“situation.”)
S4: Situation.
The first turn, I was successful if we use my getting the answer I wanted as a measure of
effectiveness. In the third turn, F, I repeated the student‟s response and started the cycle again
with a new question:
T: Situation? What kind of situation? (“troublesome” or “difficult.”)
S: Trouble? Trouble situation.
I took that to be an acceptable answer since I was not eliciting a grammatically correct
response. Focus on form vs. focus on forms is a valid argument to consider with regard to
instructional effectiveness; however it goes beyond the scope of this research.
In the next phase I tested the students‟ ability to use the new word, “predicament” in context:
T: What was the predicament? (“He spilled juice on the couch.”)
S3: Spilled the juice on the couch.
Perhaps Nunan would say that this stage of the lesson was a misuse of class time, since I
wasn‟t engaging the students in genuine communication. Seedhouse may argue that I was
using scaffolding to prepare the students to talk about predicaments. Guan Eng Ho might
agree that my questions were effective in that I was able to get the answers I was looking for.
It should be noted that most questions discussed here are display questions. Referential
questions were generally less effective than display questions:
T: What would you have done in the first situation? (grammatical focus: “would
have”)
S3: …I should have cleaned…it immediately.
Another example of an ineffective use of a referential question:
T: Would you have sent an email to apologize?
S1: Yeah, he invite…his friend…out for dinner
If we look at some display questions that were labeled “ineffective,” we can catch a glimpse
of how questions can shape the discourse of the classroom, in an attempt to steer students
toward learning objectives.
T: How did he try to solve the problem? (He turned the cushions over.)
11
S4: He would clean it.
The student‟s response was incorrect both grammatically and semantically. I followed up his
response with another display question:
T: Did he clean it?
S4: Clean? Oh. I don’t know.
In this case I would side with Seedhouse and say that, while the question itself was
ineffective at eliciting the “proper” response, the interaction was not entirely unproductive,
since the students were sending me messages about what they had and hadn‟t understood—
in this case, that I was looking for a meaning-based response, and not a grammatical one, as
can be seen in the next interaction:
T: Ok. He turned…
S2: Cushion?
This time, the third turn was not a question, but an example of corrective feedback, which
will be discussed now.
4.3 Types of feedback in my classroom
Table Two shows the types and frequency of feedback given in my class:
Recast //////////////////////// 24
Back Channeling ///////////////// 17
Repetition ////////////// 14
Praise //// 4
Clarification request // 2
Nonverbal cue / 1
Criticism / 1
Metalinguistic explanation / 1
Table 2
The quantitative results here fit current research in ELT, in that recasts were the most
common feedback type, as can be seen in the following extract:
12
S3: Trouble situation.
T: You can say troublesome situation.
Back channeling was the second most common type, and this included comments such as
„OK,‟ „mhm,‟ and „Alright.‟ Repetition was by and large used in the accuracy stage of the
lesson. When a student gave a correct answer, I simply repeated his answer before moving on
to the next student. A typical interaction during this stage of the lesson is as follows:
T: Tony? (sentence with “should have”)
S1: You should have studied very hard.
T: Alright. Should have studied very hard.
Here, the third turn begins with back channeling and finishes with repetition of the student‟s
correct answer. Another example of IRF, with repetition as the key component of F can be
seen here:
T: Wouldn’t have.
S3: I wouldn’t have gone alone.
T: OK. Wouldn’t have gone alone. Good.
I did manage to add a bit of praise with the comment, “Good.” Of the four instances of overt
praise, I used “Good” twice. Other expressions were “That‟s a very good question,” and “I
think you guys have got it.”
The lesser-used feedback strategies included a nonverbal cue, in which I mimed in order to
elicit the phrasal verb “turn over;”and criticism, where I gave general feedback to the class:
T: I think you’re having a bit of trouble producing original sentences with this
grammar.
At the beginning of class, I drew the students‟ attention to the pronunciation of “couch,”
which I labeled as “metalinguistic” feedback.
13
4.4 Effectiveness of feedback
Table Three shows the students‟ response (or lack thereof) to corrective feedback given in my
class:
Feedback Uptake
FB incorporated into next utterance ////////// 10
FB incorporated later utterance / 1
No opportunity for FB incorporation ////// 6
Ignored feedback /// 3
Table 3
I would not include back channeling in this analysis since this feedback type was used strictly
to indicate successful communication, as in this extract:
S3: You should have been more careful.
T: Mhm. Mhm.
I would also contend that repetition is not in itself corrective, and serves the same purpose of
back-channeling in that it lets the student know that his response was understood and
acceptable. Zamel would most likely disapprove of my use of repetition, as in the following
example:
S4: I would have called the hotel lobby.
T: I would have called the hotel lobby.
In order to make this third turn more effective, I should have incorporated new information
into this feedback, for example,
“I would have called the hotel lobby, or, would have talked to the front desk, or,
would have called the police.”
Of the twenty instances of correction, slightly more than half were considered “effective,” if
we use the learner‟s uptake as defined and discussed by Havranek, Mackey, and Loewen and
Philp. There was one example of feedback being incorporated into a later utterance, with
regard to the pronunciation of “couch.”
The following two examples show feedback incorporated into the learner‟s next utterance,
(immediate uptake):
14
S5: You shouldn’t have been more careful.
T: Do you mean I should have been more careful? Should have?
S5: You should have been more careful.
Here, I offered a recast, focusing on the positive form of the modal “should.” This form of
correction is more explicit than other instances of correction, though:
S5: You should have gone…gone there. Huh? You shouldn’t have…uh,…
T: Shouldn’t?
S5: You should have gone there with sun block.
In this case, intonation and stress alerted the student to his misuse of “shouldn‟t.” However, I
believe Zamel would disagree with my method here, as I ran the risk of being too vague in
my correction. Following is an instance of feedback which would likely be considered
ineffective by Zamel, Mackey and Gass:
S5: I wouldn’t have made a reservation like this hotel.
T: Or, at this hotel.
S5: Yes.
In this case, I used a recast to offer syntactical correction, and the student merely ignored my
feedback. Thus, the third turn of the IRF cycle was essentially useless. This confirms what
Havranek said about learner contribution and recasts.
In the above example, the student was given an opportunity to self-correct. However, in many
cases, the students were not given such an opportunity:
S3: …I should have cleaned it immediately.
T: So you can say I would have cleaned it immediately. Maybe, he should have
cleaned it…Jerome, what would you have done?
S1: Yeah,…
While Mackey put recasts in a positive light, the effectiveness of my feedback in this instance
is questionable at best.
15
5. Reflections
I was not surprised that the types of questions in my class are mostly display questions. This
matches the current research in ELT. It appears that I use display questions for checking
concepts and eliciting correct answers to homework. Before embarking on this research I was
not aware of the difference between display and referential questions. After looking at my
own question types, I have deduced that the purpose of my referential questions was to give
the students a chance to create using the language paradigms presented in class.
My feedback strategies are also indicative of current research in that I relied primarily on the
use of recasts. Perhaps this is an instinctive method of correction, which would explain its
prevalence in the field of ELT. I was also unaware of my frequent use of repetition as a form
of feedback. In addition, the lack of chances given to the students to self correct was
surprising.
In the future I plan to use not necessarily more or less of a particular question type, but to
make the questioning stage more communicative by giving other students a chance to give
feedback to their fellow classmates. I will also strive to give more correction in order to
promote learner uptake. Finally, I will do my best to examine the questioning and feedback
strategies I employed after each lesson, and to determine the effectiveness of those strategies.
5.1 Limitations of the study
I would like to stress again that opportunity sampling is a less-than-ideal environment for
research. The research population in this study is of negligible size and the external validity
of this study would be minimal at best. We should also remember Havranek‟s warning that
analysis of this nature would be much more valuable if research were conducted over a
longer period of time.
16
6. Conclusion
The goal of this paper was to examine my own questioning and feedback strategies and to
determine their effectiveness. I found that I used primarily display questions, the purpose
being to concept check and to check answers to homework. I used fewer referential questions,
and their purpose was generally to give the students a chance to use the new language
creatively. The effectiveness of these questions depends on the yardstick used—Seedhouse
and Nunan have conflicting views.
I am largely unsatisfied with my feedback strategies, since I relied heavily on recasts, which
are, according to many, an ineffective form of feedback. I also found that I utilize repetition
with no clear aims other than to show my understanding, which is also a form of back
channeling. I found that my students‟ uptake could have been much higher had I given them
the chance to incorporate feedback into their next utterance.
It may be useful to delve further into the purposes of display and referential questions, and to
use what was gleaned here to determine their effectiveness. It may also be interesting to
explore other categorizations of questions and determine to which end of the IRF spectrum
they belong, picking up where Van Lier left off. Additionally, the IRF cycle is a fascinating
concept and one which, in my opinion, deserves further study— particularly ways in which
the third turn can best influence the subsequent initiation phase.
17
7. References
Barnes, D. (1975) Language, The Learner, and the School. Aylsebury: Hazell
Watson and Viney.
Chauldron, C. (1977) A Descriptive Model of Discourse in the Corrective Treatment
of Learners‟ Errors. Language Learning, 27, 29-46.
Donato, R. (1994) “Collective scaffolding in second language learning.” In J.P.
Lantolf & G. Appel (ed.) Vygotskian Approaches to Second Language Research,
Norwood, NJ: Ablex. pp. 33-56.
Dornyei, Z. (2007) Research Methods in Applied Linguistics. Oxford: Oxford
University Press.
Doughty, C. (1994) “Fine-tuning of feedback by competent speakers to language
learners.” In J.E. Alatis (ed.) Georgetown University Roundtable on Languages
and Linguistics. Washington, DC: Georgetown University Press. pp. 96-108.
Ellis, R., Basturkmen, H., and Loewen, S. (2001) Learner Uptake in Communicative
ESL Lessons. Language Learning, 51, 281-318.
Guan Eng Ho, D. (2005) Why Do Teachers Ask the Questions They Ask?. RELC
Journal, 36/3, 297-310
Havranek, G. (2002) When is corrective feedback most likely to succeed?.
International Journal of Educational Research, 37, 255-70.
Loewen, S. (2002). The Occurrence and Effectiveness of Incidental Focus on Form in
Meaning-Focused ESL Lessons. Unpublished Doctoral Dissertation, The University
of Auckland, New Zealand.
Loewen, S. (2004) Uptake and Incidental Focus on Form in Meaning-Focused ESL
Lessons. Language Learning, 54, 153-87.
Loewen, S. and Philp, J. (2006) Recasts in the Adult English L2 Classroom:
Characteristics, Explicitness, and Effectiveness. The Modern Language Journal,
90/4, 536-56.
Long, M. and Sato, C. (1983) “Classroom Foreigner Talk Discourse: Forms and
Functions of Teachers.” In H.W. Seliger and M. H. Long, Classroom-Oriented
Research in Second Language Acquisition. Rowley: Newbury. pp. 268-86.
Lyster, R. (1998) Recasts, Repetition, and Ambiguity in L2 Classroom Discourse.
Studies in Second Language Acquisition, 20, 399-432.
18
Lyster, R. and Ranta, L. (1997) Corrective Feedback and Learner Uptake: Negotiation
of Form in Communicative Classrooms. Studies in Second Language Acquisition,
19, 37-66.
Mackey, A. (2006) Feedback, Noticing, and Instructed Second Language Learning.
Applied Linguistics, 27/3, 405-30.
Mackey, A. and Gass, S. (2005) Second Language Research: Methodology and
Design. London and New York: Routledge.
Nunan, D. (1987) Communicative language teaching: Making it work. ELT Journal,
41/2, 136-45.
Oliver, R. (2000) Age Differences in Negotiation and Feedback in Classroom and Pair
Work. Language Learning, 50, 119-51.
Panova, I. and Lyster, R. (2002) Patterns of Corrective Feedback and Uptake in an